The long term goal of this project is to utilize ananucleoside analogs of cytidine to understand the role of DNA methylation in cell differentiation. We have previously shown that 5-azacytidine (5-aza-CR) and 5-azadeoxycytidine (5-aza-CdR) induce profound changes in the differentiated state of mouse C3H 10T1/2 C18 cells and have attributed these changes to the abilities of the analogs to perturb DNA methylation. We now want to undertake more detailed studies on the mechanisms of 5-azanucleoside action and to use clonal derivative of 10T1/2 cells treated with these agents to understand aspects of de novo methylation and the molecular control of cell differentiation. The mechanism of cytotoxicity by 5-aza-CdR will be investigated in a series of clonal derivatives derived after multiple treatments of 10T1/2 cells with the drug. These cells are markedly resistant to the cytotoxic activity of 5-aza-CdR and will be defined with respect to gene amplification, 5-aza-CdR incorporation and the interaction of incorporated 5-azacytosine with nuclear proteins. The cells have markedly reduced levels of 5-methylcytosine (5mC) and undergo consideragle de novo methylation when dividing in the absence of further drug treatment. We will define pathways for the establishment of new methylation patterns within the cells at the level of specific gene sequences. we will also attempt to experimentally raise DNA 5mC levels within cells by treatment with inhibitors of DNA synthesis so that DNA hypermethylation can be modulated and genes may be extinguished within cells. A major part of the proposal is devoted to themolecular and cellular characterization of myogenic derivatives derived from 10T1/2 cells by 5-aza-CR treatment. We will attempt to isolate genes specific for the commitment phase of myogenic differentiation within these cells and define their methylation status and control. The hypothesis being tested here is that 5-aza-CR induces the undermethylation of one or a small number of genes whose subsequent expression defines the myogenic phenotype. Overall, these studies will increase our knowledge of the mechanisms of 5-aza-CdR action, define mechanisms for de novo methylation within cells and characterize regulatory genes in differentiating mouse cells.